I. Field
The following description relates generally to communication systems, and more particularly, to a method and apparatus for calibrating oscillators.
II. Background
In the integrated radio transceivers of many communication systems, voltage controlled oscillators (VCO) are used to generate the clock signals that are necessary for proper operation. For example, the signal from a VCO may be used to generate the radio frequency (RF) frequencies for use in frequency synthesizers. In the field of wireless communications, where a large number of channels are crowded in narrow high frequency ranges and frequency synthesizers are widely used to generate stable frequencies for transceivers, either as the transmitting frequencies or the local oscillator frequencies to beat with the receiving signals, even small variations in the operation of the VCO can cause problems.
The performance of the communication system, consequently, is critically dependent on the purity of the synthesized high-frequency output signals. For example, during signal reception, if the frequency source is impure, the result will be a mixing of undesired channels into the desired channel signal. During signal transmission, an impure frequency source will create interference in neighboring channels. Thus, it must follow that the frequency synthesizer and the underlying clock source must be carefully designed. However, even if these design concerns are addressed, variations in manufacturing processes and temperature fluctuations during operation can cause deviations in the center frequency of the VCO when it is free-running.
In view of the various unknowns and the required accuracy during normal operation, a calibration process for the VCO needs to occur at initialization of the communication system, and the clock generation system in general, is required. The wireless communication system needs to perform clock calibration so the other systems can operate. However, the time spent for calibration of the clock generation system is at a premium in wireless systems. Typically, the calibration or search is implemented using a binary search approach. This approach is the most generic of sorted search methods and does not assume prior knowledge of the system. The algorithm converges to the desired solution in a fixed number of steps, the number being proportional to the resolution desired. For example, the more accurate the refinement, the more steps will be required. However, the binary search only uses the direction of the error to get closer to the desired value.
Consequently, it would be desirable to address one or more of the deficiencies described above.